Gestión administrativa

Participants’ responses to the question “what do you understand by the term ‘web accessibility’?” varied substantially in terms of completeness and complexity. The concepts that the participants mentioned, however, were fairly consistent, suggesting a largely uniform mental model of web accessibility. As described in the previous section, five broad concept categories emerged from analysis of the participants’ responses.

• Definitions of web accessibility: statements about the meaning of the term, for example “web accessibility is about making websites accessible to people with disabilities”.

• Beneficiaries of web accessibility: descriptions of users groups who benefit from web accessibility, for example, blind or partially-sighted people.

• Technologies relating to web accessibility: descriptions of both assistive and mainstream technologies, including hardware and software, for example, screen readers or alternative keyboards and mice.

• Coding practices for implementing web accessibility: descriptions of techniques and best practices for developing accessible websites, for example, providing alternative text for images or ensuring videos have subtitles.

• Methods for evaluating web accessibility: descriptions of accessibility evaluation methods, for example, using an automatic evaluation tools or conducting user testing.

The following is a breakdown of each category.

5.4.1.1. Definitions of web accessibility

Common to each of the 26 participants’ responses was a basic definition of web accessibility. Though the complexity and completeness of their responses varied, each participant’s definition was reasonably accurate. Furthermore, though their precise terminology varied, the participants’ definitions of web accessibility were fairly consistent. Figure 5.1 provides a breakdown of the participants’ responses.

Figure 5.1: Breakdown of participants’ definitions of web accessibility

Just over half of the 26 participants (54%, 14) felt that web accessibility is a creative process, drawing upon common terms such as “making”, “creating” and “developing” in their definitions. In contrast to this, nearly a third of participants (31%, 8)

characterised it more as a process of conformance, using terms such as “making sure” and “ensuring”. 15% (4) of participants provided more unique responses, charactering web accessibility as a process of measuring or helping.

Unsurprisingly, almost all of the participants (85%, 22) considered web accessibility to be about websites, web applications or closely related concepts. Other terms used by the remaining 15% (4) of participants included: “interface”, “digital products”, and “anything that is delivered over the web”. One participant referred instead to “any vital, important content” [WD4] rather than a website or web application. Only one participant

defined web accessibility without referring to any of the above terminology, stating that it is about: “the number of people who can access something” [WD22].

Just over half of the participants (54%, 14) stated that web accessibility was about making websites or web applications “usable”, whereas just under a third of participants (31%, 8) instead used the more obvious term “accessible”. This subtle distinction may simply have arisen from participants being reluctant to provide a circular definition of web accessibility, with one noting: “it’s going to be very difficult to not say ‘accessible’!”

friendly”, “as readable as possible”, “suited to the needs” and “seen or watched or understood”.

The participants’ definitions also varied by the scope of the perceived beneficiaries of web accessibility. The most common response, provided by 42% (11) of participants, was that web accessibility is specifically for people with disabilities. 38% (10) of participants broadened the scope of web accessibility, considering it to be about

allowing as broad and diverse a range of people as possible to access or use a website or web application. 12% (3) of participants broadened the scope even further, describing web accessibility as the process of making websites accessible to everyone, irrespectively of their ability, location or technology. This inclusive approach was summarised by one developer who argued that web accessibility should include,

not only those who have significant challenges through life, more than just the web, but also those with ability and disability, essentially. Perhaps the perfect goal of 100% of anyone who needs to use a website being able to use that website [WD25].

Curiously, two web developers (8%) defined web accessibility in terms of making websites accessible to the ATs used by people with disabilities, rather than the people themselves.

Participants’ mental models of web accessibility incorporate a reasonably accurate and consistent definition of the term, characterising it as a user-centred process of

developing websites that are usable and accessible to all people, including those with disabilities. However, their choice of verb in describing the process (e.g. “making” or “creating” versus “making sure” or “ensuring”) reveals an important conceptual distinction: for some web accessibility is an act of creativity, for others it is an act of conformance.

5.4.1.2. Beneficiaries of web accessibility

As well as providing a definition of web accessibility, the majority of participants (85%, 22) also described specific user groups. Overall, the participants described a range of specific user groups who benefit from accessible websites. However, by far the most frequently mentioned user groups were people with visual disabilities and people with physical disabilities. Figure 5.2 provides a breakdown of the participants’ responses.

Figure 5.2: Beneficiaries of web accessibility

The majority (77%, 20) of the 26 participants included people with visual disabilities in their responses. Eight of these (31%) included only people with visual disabilities in their responses. 19% (5) of participants specifically mentioned blindness or not being able see at all. 15% (4) of participants specifically mentioned partially-sightedness or not being able to see very well. 27% (7) of participants included colour blindness in their responses, and 8% (2) referred to people who have poor eyesight in general.

A number of quotes from participants illustrate the predominance of people with visual disabilities in their mental models. For example, one participant said: “I suppose the main group that springs to mind is sight, so partially-sighted people, colour-blindness are the general groups really” [WD2]. Another participant said: “The people that come to

mind are blindness, maybe other visual impairments, so colour blindness, or degrees of blindness and visual impairment, e.g. bad eyesight for some reason” [WD1]. Another

participant said: “I had a few people using stuff that I've made where they were blind. I'd say also people who are colour blind. Other than that, I'm not really sure” [WD8]. One

participant stated that web accessibility is specifically for: “people that might have an impairment such as someone who is blind or someone who has… [I’m] trying to think of different impairments!” [WD11].

People with physical disabilities were the second most frequently mentioned user group and was included by just over a third (35%, 9) of participants in their responses. 19% (5) of participants specifically referred to difficulties in using a keyboard or mouse. 15% (4) of participants specifically referred to impaired motor skills. None of the participants included only people with physical disabilities in their responses. However, 27% (7) of participants included only people with visual disabilities and people with physical disabilities, with one participant describing the user groups as “the two big ones” [WD7].

The common co-occurrence of people with visual disabilities and people with physical disabilities in the participants’ mental models is illustrated by a number of quotes. For example, one participant said: “I think the main group that we always start with is the visually impaired. And then think about motor skills as well” [WD4]. Similarly, another

participant said: “when people talk about accessibility, blind people is the one that comes up the most often. People who have physical impairments when using mice and keyboards” [WD18]. Another participant stated that web accessibility was: “mostly for

sight impaired people and sometimes if you can't use your keyboard and mouse” [WD16].

Another participant stated that web accessibility was about making websites “as accessible as possible for a variety of users bearing in mind that not everybody uses a mouse, not everybody uses a keyboard, not everybody is perfectly sighted and so on”

[WD22].

Only two participants (8%) included people with hearing disabilities in their responses, one of whom stated that having a hearing impairment is: “a lesser issue” because “you don’t necessarily need to hear most things on the web” [WD9]. The other participant said:

“Obviously we barely use any Flash or anything so we don't need to worry about hearing impaired” [WD16].

Only two participants’ (8%) mental models of web accessibility include people with specific learning disabilities. Of these, one participant specifically mentioned dyslexia and the other mentioned dyslexia, dyspraxia and dyscalculia. Neither participant referred

only to people with specific learning disabilities in their responses.

Only two participants (8%) included people with age-related disabilities in their

responses, one referring specifically to visual problems encountered by older adults and the other citing older adults’ unfamiliarity with technology as opposed to an actual disability.

In accordance with the broader definitions of web accessibility provided by some participants, 19% (5) of participants stressed how everyone can benefit from web accessibility and included people with no disabilities in their responses. Of these, one participant included “end users who won’t be very technical” and “people who don’t use the language you’re presenting [the website] in” [WD15]. One participant said: “I also

consider [web accessibility] to mean things that are accessible normally, so making sure that type is big enough to read for a normal user, not necessarily someone who has a medical issue” [WD1].

Participants’ mental models of web accessibility incorporate a limited awareness of the user groups who benefit from accessible websites. Their focus centres largely on people with visual disabilities and people with physical disabilities: two user groups who

encounter very distinct and apparent challenges in accessing the web. The web is often considered to be a visual medium due to how websites are typically rendered in

graphical web browsers. This presents a number of challenges of people with visual disabilities. Similarly, the increasingly interactive nature of the web can be particularly challenging for people with physical disabilities. User groups whose access issues are less obvious (such as people with hearing disabilities) are either casually dismissed by the participants or not mentioned at all. The participants’ awareness of disabled people may be based on the extent to which people’s disabilities appear to limit access to the web. This hints at a narrow conceptual understanding of web accessibility concerned more with facilitating technical access to websites than supporting an inclusive user experience.

5.4.1.3. Technologies relating to web accessibility

As well as providing a definition of web accessibility and describing specific user groups, just over half of the participants (58%) referred to specific technologies (assistive or otherwise) in their description of web accessibility. Some participants did not mention specific technologies but described events that would imply the use of a visual display (e.g. in relation to the appearance of fonts or colour schemes). Figure 5.3 provides a breakdown of the participants’ responses.

Figure 5.3: Technologies relating to web accessibility

The most frequently mentioned technology was screen readers, mentioned by 42% (11) of participants. This is in accordance with people with visual disabilities being the most frequently mentioned user group. Interestingly, none of the web developers referred to any specific screen reader software (e.g. JAWS or VoiceOver).

The prominence of screen readers in the participants’ mental models is illustrated in a number of quotes. For example, one participant said: “accessibility involves making sure your website can be browsed by different types of user agents, e.g. a screen reader for people with a visual impairment” [WD1]. Another said web accessibility: “is about getting

technology right, so it's things like making sure content is available to screen readers or content locked away in proprietary technologies like Flash is available as plain text. Doing it wrong is not doing that stuff” [WD14].

Six participants (23%) acknowledged the need for websites to work across a range of input devices and user agents and to be “technology neutral”. This was not limited only to ATs, with two participants referring specifically to keyboards, and another to mobile phones.

The participants’ emphasis on providing compatibility across different technologies is illustrated by a number of quotes. For example, one participant said: “If I was using a mouse, I might expect someone with impaired motor skills to not have access to a

mouse and be using another input device” [WD10]. One participant stated that web

accessibility is: “generally just making the website vaguely friendly to screen readers and technologies to actually amplify the web experience for impaired people” [WD16]. Another

said: “You're making sure that things are keyboard accessible and accessible to screen readers and accessible to people who are visually impaired or colour-blindness or whatever” [WD13]. Another participant felt that web accessibility is “the idea of

semantically building sites such that they are more accessible for different tools, different systems and ways of using the web” [WD8].

Two participants (8%) felt that it was important to consider people who might be using older technology, with one participant citing the example of “someone who is in a poorer country who might not necessarily be using a massive great screen with the very latest browser” [WD24]. The other participant argued that the only technological

requirement to access the web should be a computer and an Internet connection, adding: “if you have access to those, [web accessibility is] this egalitarian idea that anyone with access to those basic technologies should, as far as possible, be equally able to use the services that are available through them” [WD25].

Participants’ mental models of web accessibility incorporate a limited awareness of the ATs used by people with disabilities. Focusing mainly on screen readers and alternative input devices, their knowledge of ATs appears to reflect their limited awareness of disabled people. What is particularly revealing, however, is that many participants describe web accessibility in terms of supporting different user agents but not in terms of supporting actual users. This, again, suggests a conceptual understanding of web accessibility that prioritises technical requirements over user needs.

5.4.1.4. Coding practices for implementing web accessibility

Two participants (8%) felt that web accessibility does not require any particular accessible coding practices. One argued that web accessibility is “generally covered by the way you build it anyway” [WD4]. The other stated that that to create an accessible

website you “don’t need anything, just programming” [WD21]. However, a large

proportion (69%, 18) of participants described specific coding practices for creating accessible websites. Figure 5.4 provides a breakdown of the participants’ responses.

Figure 5.4: Coding practices for implementing web accessibility

Over a third of the participants (38%, 10) mentioned coding practices to support people with visual disabilities. These were further categorised into coding practices that support people that have some sight (mentioned by 31% (8) of participants) and coding

practices that support people with no sight at all (mentioned by 23% (6) of participants).

Coding practices that support people with some sight were categorised further still into coding practices related to colours (mentioned by 27% (7) of participants) and coding practices relating to fonts (mentioned by 12% (3) of participants). The coding practices relating to colours included: providing appropriate colour schemes; ensuring webpages have sufficient colour contrast; and avoiding relying on colour to convey information. The coding practices relating to fonts included: making fonts large enough to read; and providing a means of allowing users to adjust the font size.

One participant said: “For me [web accessibility] is mostly stuff like... colours on screen is a big concern, so thinking about how things display to people with different types of colour blindness” [WD7]. In contrast, one participant felt that colour scheme decisions

were the responsibility of a web developer, arguing: “I suppose the designer would design it using colours that aren't unsuitable for colour blind people. Obviously, that's more the designer's responsibility so not greens and blues or whatever” [WD2].

Coding practices that support people with no sight at all were categorised further still into coding practices relating to text alternatives (mentioned by 15% (4) of participants) and coding practices relating to page navigation (mentioned by 8% (2) of participants). The coding practices relating to text alternatives included: providing text alternatives for images; and providing text alternatives for Flash components. The coding practices relating to page navigation included: providing a meaningful order of information; and providing skip links at the top of the webpage.

Participants frequently described their efforts in terms of supporting ATs as opposed to enabling users. One participant said: “I try and follow what I believe to currently be best practices, in terms of ordering things so that screen readers can pull out the most important information” [WD20]. Another said:

I think screen readers look for different tags and that kind of thing and structure and conforming to that helps make things better. One of the simplest things is if semantically your navigation is at the top of the code, then having an invisible link that says ‘skip to content’ so that screen readers see that first, and you can use that to skip to content so you don't have the whole navigation read out [WD8].

12% (3) of participants suggested coding practices to support people with physical disabilities. These were further categorised into coding practices relating to buttons (mentioned by 8% (2) of participants) and coding practices relating to positioning (mentioned by 4% (1) of participants). The coding practices relating to buttons included making buttons sufficiently large enough. The coding practices relating to positioning included positioning elements to minimise mouse movements.

One participant stated that web accessibility involved: “making sure buttons aren't too fiddly and thinking about how you could operate a site without using a mouse” [WD4].

Another said: “If you've got to do some arduous task to move the mouse across the screen you might want things like being able to use the tab key, rather than having to click in every box” [WD9]. One participant argued that button design is not the

responsibility of the web developer, stating: “it is more the responsibility of a designer to make sure buttons are big enough to be visible” [WD2].

Another set of coding practices suggested by 19% (5) of participants did not relate to a particular user group but instead to support web accessibility in general. Each of these coding practices related to the use of web technologies, including: using semantic HTML; using CSS; using web development best practices; and ensuring webpages conform to web standards.

The prominence of standards conformance in the participants’ mental models is illustrated in a number of quotes. One participant said web accessibility involves,

making sure that [a webpage] is actually sensibly marked up. It actually meets standards, that you don't abuse tags, like put everything in paragraph tags for example. Actually mark it up so it can be read without the style sheet or anything attached to it [WD16].

According to another participant, web accessibility “is a wide-ranged approach depending on how you write your markup, whether that's semantic” [WD26]. One

participant stated:

As a developer I would have two definitions [of web accessibility]: one is more related to me, which is web standards: does the website validate correctly, are you using correct HTML and all that stuff. And the other one is can people use it effectively [WD2].

Participants’ mental models of web accessibility incorporate a limited procedural understanding of how to implement accessible websites, which reflects their limited knowledge both of disabled people and the ATs they use. Despite referring to a range of accessible coding practices in their responses, participants appear to be most familiar with those that support people with visual disabilities and people with physical

disabilities. Notably, participants describe coding websites in a particular way to support screen readers and ATs rather than the people who use them. This again indicates a conceptual understanding of web accessibility that focuses more on technical access